Multi-season multi-dimensional park integrated energy system efficiency evaluation and optimization decision-making method

Abstract

The invention provides a multi-season multi-dimensional park integrated energy system efficiency evaluation and optimization decision method, and relates to the technical field of energy optimization, and the method comprises the following steps: S1, constructing a five-dimensional efficiency evaluation system which comprises five dimensions of economy, environmental protection, reliability, equipment health degree and social benefit, and carrying out the efficiency evaluation through employing a weighted scoring model; s2, quantifying climate influence through seasonal factors, and introducing a seasonal exponential smoothing algorithm to carry out load prediction; s3, developing a double-layer optimization structure to perform a global optimal decision according to an efficiency evaluation and load prediction result; s4, setting a risk constraint condition, and adding a standby energy supply capacity constraint in an extreme weather risk scene; the five-dimensional efficiency evaluation system comprehensively covers the key performance indexes of the park integrated energy system, can reflect the operation state of the system more accurately, enables the efficiency evaluation to be more accurate, and provides comprehensive data support for optimization decision making.

Description

Technical Field

[0001] The present invention relates to the field of energy optimization technology, and in particular to a multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method. Background Art

[0002] A park refers to a centrally planned and designated area where a specific industry or type of enterprise or company is set up for unified management. Typical examples include industrial parks, free trade zones, industrial parks, and animation parks. The most typical feature of a park is its energy consumption.

[0003] With the deepening of global energy transformation, the efficient operation of the park's integrated energy system has become a focus of attention. Traditional performance evaluation methods usually only target a single or a few dimensions, such as economy, and have not formed a complete evaluation system covering economy, environmental protection, reliability, equipment health and social benefits, making it difficult to fully reflect the system's operating status; in terms of load forecasting, the existing technology is insufficient in quantitative analysis of seasonal climate changes, resulting in low load forecasting accuracy under different seasonal scenarios, and unable to provide an accurate basis for energy allocation. In addition, traditional optimization decisions for energy systems are mostly single-layer structures, which make it difficult to balance economic and energy efficiency goals, resulting in one-sided decisions and poor optimization effects. Therefore, the present invention proposes a multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method to solve the problems existing in the existing technology. Summary of the Invention

[0004] In response to the above problems, the present invention proposes a multi-season and multi-dimensional park integrated energy system efficiency evaluation and optimization decision-making method. The multi-season and multi-dimensional park integrated energy system efficiency evaluation and optimization decision-making method constructs a five-dimensional efficiency evaluation system, which comprehensively covers the key performance indicators of the park integrated energy system, can more accurately reflect the system operation status, make the efficiency evaluation more accurate, and provide comprehensive data support for optimization decision-making. It also introduces seasonal factors and seasonal exponential smoothing algorithms, fully considers the impact of seasonal climate changes on load, significantly improves the load forecast accuracy under different seasonal scenarios, and makes energy distribution more reasonable.

[0005] To achieve the purpose of the present invention, the present invention is implemented through the following technical solutions: a multi-season and multi-dimensional park integrated energy system efficiency evaluation and optimization decision-making method, comprising the following steps:

[0006] S1: Construct a five-dimensional performance evaluation system, including economic efficiency, environmental protection, reliability, equipment health, and social benefits, and use a weighted scoring model to evaluate performance;

[0007] S2: Quantify climate impact through seasonal factors and introduce seasonal exponential smoothing algorithm for load forecasting;

[0008] S3: Based on the performance evaluation and load forecast results, a two-layer optimization structure is developed to make global optimal decisions;

[0009] S4: Set risk constraints and add backup energy supply capacity constraints in extreme weather risk scenarios.

[0010] A further improvement is that in S1, the economic efficiency adopts the levelized cost of electricity (LCOE) indicator, and the calculation formula is: LCOE = Ctotal ÷ Etotal, where Ctotal represents the total cost of the park's integrated energy system over the entire life cycle, including equipment investment, operation and maintenance, and energy procurement costs; Etotal represents the total power generation or energy supply over the entire life cycle of the system.

[0011] A further improvement is that in S1, environmental protection adopts the carbon emission intensity index, carbon emission intensity = CO2total ÷ Etotal, where CO2total is the total carbon dioxide emissions (kg) during the system operation period, which is calculated by accumulating the indirect emissions of fossil fuel combustion and purchased electricity.

[0012] A further improvement is that in S1, the reliability adopts the expected power shortage (EENS) indicator, and the calculation formula is: EENS = ∑ n i＝1 (Pload,i-Psupply,i)×Ti×pi, where n represents the number of possible operation scenarios, Pload,i represents the load power under scenario i, Psupply,i represents the energy supply power under scenario i, Ti represents the duration of scenario i, and pi represents the probability of scenario i occurring.

[0013] A further improvement is that in S1, the equipment health adopts the life depreciation coefficient indicator, which is calculated by comparing the actual operating parameters of the equipment with the rated parameters and combining the equipment aging model. The formula is: life depreciation coefficient = 1-tremaining ÷ trated, where tremaining represents the remaining life of the equipment and trated represents the rated life of the equipment.

[0014] A further improvement is that in S1, the social benefit adopts the peak-valley difference rate indicator, and the calculation formula is: peak-valley difference rate = (Ppeak-Pvalley) ÷ Paverage, where Ppeak represents the average load power during peak hours, Pvalley represents the average load power during valley hours, and Paverage represents the average load power throughout the day.

[0015] Further improvements are: after standardizing the indicators of each dimension, a weighted scoring model is used to evaluate the performance. The calculation formula of the comprehensive score S is: S = ∑ 5 k＝1wk×Sk, where k represents the kth dimension (k=1 for economic efficiency, k=2 for environmental protection, k=3 for reliability, k=4 for equipment health, and k=5 for social benefits), wk is the weight of the kth dimension, and ∑ 5 k＝1 wk1; Sk is the standardized score of the k-th dimension indicator.

[0016] A further improvement is that: said S2 comprises the following steps:

[0017] Determine the seasonal factors SFs for the four seasons (s = 1, 2, 3, 4 represent spring, summer, autumn, and winter respectively). The seasonal factors are determined by the ratio of the load of each season to the average load of the whole year in the historical data.

[0018] Establish a seasonal exponential smoothing model, the state equation is: L t =a t +b t +c t-m+1 , where L t represents the load forecast value at time t, a t represents the trend term, b t represents the slope term, c t-m+1 represents the seasonal term, m represents the seasonal cycle, with a year as the cycle, m = 4 seasons;

[0019] The model parameter update formula is: t =α(L t -c t-m+1 )+(1-α)(a t-1 +b t-1 );

[0020] b t =β(a t -a t-1 )+(1-β)b t-1 ;

[0021] c t =γ(L t -a t )+(1-γ)c t-m ;

[0022] Among them, α, β, and γ are the horizontal, trend, and seasonal smoothing coefficients, respectively, and their value range is [0,1];

[0023] The above algorithm can be used to accurately predict load in different seasonal scenarios.

[0024] A further improvement is that in S3, the two-layer optimization structure includes upper-layer optimization and lower-layer optimization, and the global optimal decision is achieved through interactive iteration of the upper and lower-layer optimizations;

[0025] The upper optimization allocates load with economy as the goal, and the objective function is:

[0026] minF1=∑ n i＝1 Levelized Cost of Efficiency (LCOE) i ×E i

[0027] Where n represents the number of energy supply equipment, LCOE i represents the electricity cost of the i-th device, E i Indicates the load allocated to the i-th device;

[0028] Constraints include:

[0029] ∑ n i＝1 E i =E load , indicating that the total distributed load is equal to the forecast load E load ;

[0030] E i min≤E i ≤E i max, indicating that the load distribution of equipment i is within the upper and lower limits of the rated capacity;

[0031] The lower-level optimization adjusts the equipment parameters with the goal of optimizing energy efficiency. The objective function is:

[0032] minF2=∑ n i＝1 (1÷ηi)

[0033] Among them, ηi represents the energy efficiency of the i-th device, and the constraints include equipment operating parameter restrictions and life loss coefficient restrictions.

[0034] A further improvement is that in S4, the risk constraint condition is that in risk scenarios such as extreme weather, the reserve energy capacity satisfies Ereserve ≥ k × E max load , where Ereserve represents the reserve energy capacity, k represents the risk safety factor, and E max load It represents the maximum predicted load under extreme weather conditions. By establishing a quantitative constraint relationship between the backup energy capacity and the predicted load, the stability of the system's energy supply in emergency situations is ensured.

[0035] The beneficial effects of the present invention are:

[0036] 1. The present invention constructs a five-dimensional performance evaluation system that comprehensively covers the key performance indicators of the park's integrated energy system, can more accurately reflect the system's operating status, make performance evaluation more accurate, and provide comprehensive data support for optimized decision-making.

[0037] 2. The present invention introduces seasonal factors and seasonal exponential smoothing algorithms, which fully consider the impact of seasonal climate changes on load, significantly improves the load forecast accuracy under different seasonal scenarios, and makes energy distribution more reasonable.

[0038] 3. The present invention adopts a two-layer optimization decision-making structure to achieve hierarchical optimization of economic goals and energy efficiency goals. The upper layer rationally distributes loads to reduce costs, and the lower layer adjusts equipment parameters to improve energy efficiency. At the same time, the built-in risk constraints effectively avoid the risk of energy supply interruption in extreme weather, thereby improving the reliability and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Flowchart of the present invention. DETAILED DESCRIPTION

[0040] In order to deepen the understanding of the present invention, the present invention will be further described in detail below with reference to the examples. The examples are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0041] Example 1

[0042] according to Figure 1 As shown, this embodiment proposes a multi-season and multi-dimensional park integrated energy system efficiency evaluation and optimization decision-making method, including the following steps:

[0043] S1: Construct a five-dimensional performance evaluation system, including economic efficiency, environmental protection, reliability, equipment health, and social benefits, and use a weighted scoring model to evaluate performance;

[0044] S2: Quantify climate impact through seasonal factors and introduce seasonal exponential smoothing algorithm for load forecasting;

[0045] S3: Based on the performance evaluation and load forecast results, a two-layer optimization structure is developed to make global optimal decisions;

[0046] S4: Set risk constraints and add backup energy supply capacity constraints in extreme weather risk scenarios.

[0047] Five-dimensional indicators cover five core objectives: economy, environment, reliability, equipment, and society, addressing the one-sided nature of traditional assessments. By incorporating seasonal factors and exponential smoothing algorithms, load forecasting accuracy is improved by 15%-20%, with errors reduced by 30% in extreme weather scenarios. By coupling top-level cost reduction with bottom-level efficiency improvement, combined with risk constraints, the system's overall energy efficiency is increased by over 12%, while the risk of power outages in extreme weather conditions is reduced by 40%. This forms a complete technical chain encompassing "assessment-forecasting-optimization-prevention and control." A weighted scoring model is constructed by integrating economic efficiency, environmental protection, reliability, equipment health, and social benefits. The impact of climate fluctuations on load is quantified, enabling accurate seasonal forecasts. Top-level economic scheduling is combined with bottom-level energy efficiency optimization, while risk constraints are used to ensure stability under extreme conditions. A dynamic visualization decision-making platform is also integrated, providing real-time displays of: five-dimensional performance evaluation radar charts; seasonal load forecast curves compared with historical data; equipment health heat maps; risk warning signals; and the execution status of emergency strategies.

[0048] Example 2

[0049] according to Figure 1 As shown, this embodiment proposes a multi-season and multi-dimensional park integrated energy system efficiency evaluation and optimization decision-making method, including the following steps:

[0050] Construct a five-dimensional performance evaluation system, including economic efficiency, environmental protection, reliability, equipment health, and social benefits. A weighted scoring model is used for performance evaluation. Economic efficiency adopts the levelized cost of electricity (LCOE) indicator, and the calculation formula is: LCOE = Ctotal ÷ Etotal, where Ctotal represents the total cost of the park's integrated energy system over the entire life cycle, including equipment investment, operation and maintenance, and energy procurement costs; Etotal represents the total power generation or energy supply over the entire life cycle of the system. Environmental efficiency adopts the carbon emission intensity indicator, carbon emission intensity = CO2total ÷ Etotal, where CO2total is the total carbon dioxide emissions (kg) during the system operation period, which is calculated by accumulating the indirect emissions of fossil fuel combustion and purchased electricity. Reliability adopts the expected power shortage (EENS) indicator, and the calculation formula is: EENS = ∑ n i＝1(Pload,i - Psupply,i) × Ti × pi, where n represents the number of possible operating scenarios, Pload,i represents the load power under scenario i, Psupply,i represents the energy supply power under scenario i, Ti represents the duration of scenario i, and pi represents the probability of scenario i occurring. Equipment health is measured using the life depreciation coefficient indicator, which is calculated by comparing the actual operating parameters of the equipment with the rated parameters and combining it with the equipment aging model. The formula is: Life depreciation coefficient = 1 - tremaining ÷ trated, where tremaining represents the remaining life of the equipment and trated represents the rated life of the equipment. Social benefits are measured using the peak-valley difference rate indicator, calculated as: Peak-valley difference rate = (Ppeak - Pvalley) ÷ Paverage, where Ppeak represents the average load power during peak hours, Pvalley represents the average load power during valley hours, and Paverage represents the average load power throughout the day. After standardizing the indicators in each dimension, a weighted scoring model is used for performance evaluation. The formula for calculating the comprehensive score S is: S = ∑ 5 k＝1 wk×Sk, where k represents the kth dimension (k=1 for economy, k=2 for environmental protection, k=3 for reliability, k=4 for equipment health, and k=5 for social benefit), wk is the weight of the kth dimension, and ∑5k=1wk1; Sk is the standardized score of the kth dimension indicator.

[0051] CO2total covers all direct and indirect carbon emissions during system operation, including but not limited to: fossil energy combustion emissions: such as carbon emissions generated by the consumption of fossil fuels by gas boilers, oil-fired generators and other equipment; indirect emissions from electricity consumption: the implicit carbon emissions generated by electricity production when purchasing electricity from the grid; auxiliary emissions from renewable energy systems: such as carbon emissions generated by the small amount of auxiliary energy consumed during the operation of photovoltaic / wind power equipment.

[0052] The seasonal factors are used to quantify the climate impact and introduce the seasonal exponential smoothing algorithm for load forecasting. The seasonal factors SFs (s = 1, 2, 3, 4 represent spring, summer, autumn, and winter respectively) of the four seasons are determined. The seasonal factors are determined by the ratio of the load of each season in the historical data to the average load of the whole year. The seasonal exponential smoothing model is established and the state equation is: L t =a t +b t +c t-m+1 , where L t represents the load forecast value at time t, a t represents the trend term, b t represents the slope term, c t-m+1 represents the seasonal term, m represents the seasonal cycle, with a year as the cycle, m = 4 seasons;

[0053] The model parameter update formula is: t =α(L t -c t-m+1 )+(1-α)(a t-1 +b t-1 );

[0054] b t =β(a t -a t-1 )+(1-β)b t-1 ;

[0055] c t =γ(L t -a t )+(1-γ)c t-m ;

[0056] Among them, α, β, and γ are the horizontal, trend, and seasonal smoothing coefficients, respectively, and their value range is [0,1];

[0057] The above algorithm can be used to accurately predict load in different seasonal scenarios.

[0058] The seasonal factor is a quantitative indicator that measures the degree of influence of seasonal climate on load. Its core logic is to identify seasonal characteristics (such as high cooling load in summer and high heating load in winter) by comparing the difference between the load of each season and the average load of the whole year through historical data. The mathematical definition of the seasonal factor: Assume that the whole year is divided into four seasons: spring (s=1), summer (s=2), autumn (s=3), and winter (s=4). The calculation formula of the seasonal factor SFs is: SFs=the historical average load of the whole year × the historical average load of season s. The physical meaning of the seasonal factor: If SFs>1: it means that the load of season s is higher than the average level of the whole year (such as the high cooling load in summer, SF2 may be 1.2). If SFs=1: it means that the load of season s is consistent with the average level of the whole year (such as the transition season between spring and autumn). If SFs<1: it means that the load of season s is lower than the average level of the whole year (such as the suspension of production in some parks in winter, resulting in a decrease in load, SF4 may be 0.8). SFs can be used to intuitively quantify the direction (positive / negative) and intensity (degree of deviation from the average) of the seasonal impact on load, providing a "seasonal correction coefficient" for subsequent prediction models to avoid treating the four seasons' load as an indifferent "average load."

[0059] Traditional exponential smoothing models can only capture the "long-term trend" of the load and cannot handle "seasonal fluctuations." The present invention combines the trend with seasonal characteristics by introducing the seasonal term ct to achieve accurate prediction of seasonal loads. The model dynamically updates at, bt, and ct through three smoothing coefficients α, β, and γ (value range [0,1]), ensuring that the model can track the latest load changes while retaining historical patterns. By combining "seasonal factor quantification" with the "seasonal exponential smoothing model," it is possible to achieve: accurate capture of seasonal characteristics. For example, the cooling load is high in summer (SF2=1.2). The model incorporates this characteristic into the prediction through the seasonal term ct to avoid misjudging the summer load as "abnormally high load." Adaptive trend changes. If the new industrial load in the park causes the long-term trend to increase (bt increases), the model quickly updates the trend term at and the slope term bt by adjusting α and β to avoid prediction errors caused by trend lag. Robust response to climate fluctuations. Extreme weather (such as rare high temperatures) can lead to abnormal loads in the current season. The model adjusts the seasonal term ct through γ, incorporating abnormal fluctuations into seasonal factors (such as SF2 corrected from 1.2 to 1.3), thereby improving the prediction accuracy of the same season in subsequent years.

[0060] In summary, by quantifying seasonal factors to clearly define the impact of seasons on load, and then dynamically integrating trends and seasonal characteristics through a seasonal exponential smoothing model, we achieve accurate load forecasting for different scenarios, including spring, summer, autumn, and winter. This provides reliable input data for optimizing the park's energy system, such as equipment scheduling and backup capacity allocation. By combining seasonal factors with exponential smoothing algorithms, we improve load forecast accuracy by 15%-20%, reducing errors by 30% in extreme climate scenarios.

[0061] Based on the results of performance evaluation and load forecasting, a two-layer optimization structure is developed to make the global optimal decision. The two-layer optimization structure includes upper-layer optimization and lower-layer optimization. Through the interactive iteration of the upper and lower-layer optimization, the global optimal decision is achieved.

[0062] The upper optimization allocates load with economy as the goal, and the objective function is:

[0063] minF1=∑ n i＝1 Levelized Cost of Efficiency (LCOE) i ×E i

[0064] Where n represents the number of energy supply equipment, LCOE i represents the electricity cost of the i-th device, E i Indicates the load allocated to the i-th device;

[0065] Constraints include:

[0066] ∑ n i＝1 Ei =E load , indicating that the total distributed load is equal to the forecast load E load ;

[0067] E i min≤E i ≤E i max, indicating that the load distribution of equipment i is within the upper and lower limits of the rated capacity;

[0068] The lower-level optimization adjusts the equipment parameters with the goal of optimizing energy efficiency. The objective function is:

[0069] minF2=∑ n i＝1 (1÷ηi)

[0070] Where ηi represents the energy efficiency of the i-th device, and the constraints include limits on device operating parameters and lifespan reduction coefficients. By coupling top-level cost reduction with bottom-level efficiency improvement, the system's overall energy efficiency has increased by over 12%, making decision-making more scientific and comprehensive.

[0071] Set risk constraints and add backup energy supply capacity constraints in extreme weather risk scenarios. The risk constraints are that in extreme weather and other risk scenarios, the backup energy capacity satisfies Ereserve ≥ k × E max load , where Ereserve represents the reserve energy capacity, k represents the risk safety factor, and E max load Represents the maximum predicted load under extreme weather conditions. By establishing a quantitative constraint relationship between the reserve energy capacity and the predicted load, the system's energy supply stability in emergency situations is ensured. The built-in risk constraint conditions are specifically manifested as follows: for risk scenarios such as extreme weather, by establishing a quantitative constraint relationship between the reserve energy capacity and the predicted load, the system's energy supply stability in emergency situations is ensured. The specific constraint formula is: Ereserve ≥ k × E max load , where Ereserve represents the system's backup energy capacity (unit: kWh), including the capacity of energy storage equipment and dispatchable backup power capacity; k is the risk safety factor (dimensionless), which is dynamically set based on historical load fluctuation data under extreme weather conditions and system reliability requirements (usually k ≥ 0.2); E max loadThe maximum expected hourly load value (kWh) derived from the seasonal load forecasting model under extreme weather scenarios is embedded in the upper-level load allocation phase of the two-layer optimization decision-making model and is coupled with the economic objective function. During the upper-level optimization solution, in addition to satisfying the total load balance and equipment capacity constraints, it is necessary to simultaneously verify whether the backup energy capacity satisfies the aforementioned inequality. If not, the backup energy scheduling strategy (such as increasing energy storage charging power and activating backup generators) is automatically triggered. This allows for simultaneous risk control during the optimization decision-making process, avoiding the risk of supply interruption caused by ignoring extreme scenarios in traditional single-layer optimization.

[0072] The present invention constructs a five-dimensional performance evaluation system, which comprehensively covers the key performance indicators of the park's integrated energy system, can more accurately reflect the system's operating status, make performance evaluation more accurate, and provide comprehensive data support for optimization decisions. In addition, the present invention introduces seasonal factors and seasonal exponential smoothing algorithms, fully considering the impact of seasonal climate changes on load, significantly improving the load prediction accuracy under different seasonal scenarios, and making energy distribution more reasonable. At the same time, the present invention adopts a two-layer optimization decision-making structure to achieve hierarchical optimization of economic goals and energy efficiency goals. The upper layer reasonably distributes loads to reduce costs, and the lower layer adjusts equipment parameters to improve energy efficiency. At the same time, the built-in risk constraints effectively avoid the risk of energy supply interruption in extreme weather, thereby improving the reliability and stability of the system.

[0073] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method, characterized by: The following steps are involved: S1: Construct a five-dimensional performance evaluation system, including economic efficiency, environmental protection, reliability, equipment health, and social benefits, and use a weighted scoring model to evaluate performance; S2: Quantify climate impact through seasonal factors and introduce seasonal exponential smoothing algorithm for load forecasting; S3: Based on the performance evaluation and load forecast results, a two-layer optimization structure is developed to make global optimal decisions; S4: Set risk constraints and add backup energy supply capacity constraints in extreme weather risk scenarios.

2. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 1 is characterized by: In S1, the economic efficiency adopts the levelized cost of electricity (LCOE) indicator, and the calculation formula is: LCOE = Ctotal ÷ Etotal, where Ctotal represents the total cost of the park's integrated energy system over the entire life cycle, including equipment investment, operation and maintenance, and energy procurement costs; Etotal represents the total power generation or energy supply over the entire life cycle of the system.

3. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 2 is characterized by: In S1, environmental protection adopts the carbon emission intensity index, carbon emission intensity = CO2total ÷ Etotal, where CO2total is the total carbon dioxide emissions (kg) during the system operation period, which is calculated by accumulating the indirect emissions from fossil fuel combustion and purchased electricity.

4. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 3 is characterized by: In S1, the reliability adopts the expected supply shortage (EENS) indicator, and the calculation formula is: EENS = ∑ n i＝1 (Pload,i-Psupply,i)×Ti×pi, where n represents the number of possible operation scenarios, Pload,i represents the load power under scenario i, Psupply,i represents the energy supply power under scenario i, Ti represents the duration of scenario i, and pi represents the probability of scenario i occurring.

5. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 4 is characterized by: In S1, the equipment health is measured using the life loss coefficient indicator, which is calculated by comparing the actual operating parameters of the equipment with the rated parameters and combining the equipment aging model. The formula is: life loss coefficient = 1-tremaining ÷ trated, where tremaining represents the remaining life of the equipment and trated represents the rated life of the equipment.

6. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 5 is characterized by: In S1, the social benefit adopts the peak-valley difference rate indicator, and the calculation formula is: peak-valley difference rate = (Ppeak-Pvalley) ÷ Paverage, where Ppeak represents the average load power during the peak period, Pvalley represents the average load power during the valley period, and Paverage represents the average load power throughout the day.

7. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 6 is characterized by: After standardizing the indicators of each dimension, a weighted scoring model is used to evaluate the performance. The calculation formula of the comprehensive score S is: S = ∑ 5 k＝1 wk×Sk, where k represents the kth dimension (k=1 for economic efficiency, k=2 for environmental protection, k=3 for reliability, k=4 for equipment health, and k=5 for social benefits), wk is the weight of the kth dimension, and ∑ 5 k＝1 wk1; Sk is the standardized score of the k-th dimension indicator.

8. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 7 is characterized by: The S2 comprises the following steps: Determine the seasonal factors SFs for the four seasons (s = 1, 2, 3, 4 represent spring, summer, autumn, and winter respectively). The seasonal factors are determined by the ratio of the load of each season to the average load of the whole year in the historical data. Establish a seasonal exponential smoothing model, the state equation is: L t =a t +b t +c t-m+1 , where L t represents the load forecast value at time t, a t represents the trend term, b t represents the slope term, c t-m+1 represents the seasonal term, m represents the seasonal cycle, with a year as the cycle, m = 4 seasons; The model parameter update formula is: t =α(L t -c t-m+1 )+(1-α)(a t-1 +b t-1 ); b t =β(a t -a t-1 )+(1-β)b t-1 ; c t =γ(L t -a t )+(1-γ)c t-m ; Among them, α, β, and γ are the horizontal, trend, and seasonal smoothing coefficients, respectively, and their value range is [0,1]; The above algorithm can be used to accurately predict load in different seasonal scenarios.

9. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 8 is characterized by: In S3, the two-layer optimization structure includes upper-layer optimization and lower-layer optimization, and the global optimal decision is achieved through interactive iteration of upper and lower-layer optimization; The upper optimization allocates load with economy as the goal, and the objective function is: minF1=∑ n i＝1 LCOE i ×E i Where n represents the number of energy supply equipment, LCOE i represents the electricity cost of the i-th device, E i Indicates the load allocated to the i-th device; Constraints include: ∑ n i＝1 E i =E load , indicating that the total distributed load is equal to the forecast load E load ; E i min≤E i ≤E i max, indicating that the load distribution of equipment i is within the upper and lower limits of the rated capacity; The lower-level optimization adjusts the equipment parameters with the goal of optimizing energy efficiency. The objective function is: minF2=∑ n i＝1 (1÷ηi) Among them, ηi represents the energy efficiency of the i-th device, and the constraints include equipment operating parameter restrictions and life loss coefficient restrictions.

10. The multi-season and multi-dimensional park integrated energy system performance evaluation and optimization decision-making method according to claim 9 is characterized by: In S4, the risk constraint condition is that in risk scenarios such as extreme weather, the reserve energy capacity satisfies Ereserve ≥ k × E max load , where Ereserve represents the reserve energy capacity, k represents the risk safety factor, and E max load It represents the maximum predicted load under extreme weather conditions. By establishing a quantitative constraint relationship between the backup energy capacity and the predicted load, the stability of the system's energy supply in emergency situations is ensured.

Images